Traffic engineering
The general formulas used to populate the COI matrix entries in Table 26, Table 27, and Table 28, respectively, for the Uniform Distribution model are:
Inbound CUR to Site i
Outbound CUR from Site
= | ⎛ number of stations in Site i ⎞ | ⋅ (total inbound CUR) |
⎝ |
i = | ⎛ number of stations in Site i ⎞ | ⋅ (total outbound CUR) |
⎝ |
Intercom CUR from Site i to Site j = | ⎛ number of stations in Site j ⎞ | ⎛ total intercom CUR⎞ |
⎝ | ⋅ ⎝originating in Site i ⎠ |
Additional comments regarding Example 2: Uniform Distribution model
In the Uniform Distribution model introduced in Example 2: Uniform Distribution model on page 186, the relative weights that are associated with the various sites correspond to the distribution of stations throughout the sites. Alternatively, the weights could correspond to the relative overall station usages in the various sites. Such a model takes into account not only the number of stations, but also how busy the stations are. In Example 2: Uniform Distribution model, since every station is assumed to have the same usage (specifically, 0.11 Erlangs), the weights that are based on the number of stations per site are exactly the same as the weights that are based on the overall station usage per site. Such a model is not always appropriate. For example, consider a system with two sites, with 100 stations in each site. Suppose that the average usage per station in Site 1 is 0.1 Erlangs, and that the average usage per station in Site 2 is 0.2 Erlangs. In a Uniform Distribution model where the weights are based on station usage per endpoint, a caller in Site 1 is twice as likely to call a station in Site 2 than a station in Site 1 (because the total station usage in Site 2 is 20 Erlangs, and the total station usage in Site 1 is only 10 Erlangs). The general formulas used to populate the COI matrix entries in Table 26,
